Best Linux for Bioinformatics
Importance of Linux in bioinformatics
I’ve been wondering about the best operating system for bioinformatics for a long time (here I can see that I studied Computer Engineering). And if I’m honest, all I’ve gotten clear is that it’s not Windows 🙂 This is the first in a series of articles on operating systems and bioinformatics, and I’ll try to mention the most typical systems.
Many of us are using Linux distributions (Debian / Ubuntu / etc, RedHat / Fedora, SuSE / OpenSuSE, Mandrake / Mandriva, Gentoo, etc …), Unix (FreeBSD, Solaris / OpenSolaris, IRIX, Tru64, HP / UX, AIX, …), Mac OS X, or even Windows! And all of us, to a greater or lesser extent, have suffered from the following problems:
After reading an article, we wanted to install a program or library (for example, t-coffee or BioPerl), but since there are no installable packages for our system, we had to compile them by hand.
As these programs and libraries often depend on packages that are not in our distribution / operating system, we have also compiled them!
Time passes, and we wanted to update the operating system, and then, everything goes to waste! Start again!
A couple of years ago I was already totally tired of compiling, for all the time I lost looking for what I had to install, testing, etc … Also, I was also tired of every time I updated the operating system; problems could arise between what I compiled and what is updated. At that point, I thought of using some operating system with continuous updates, like some variant of Debian Linux, FreeBSD or Gentoo Linux (the latter two for radicals among radicals). Out of curiosity, I started to see what packages were available in Gentoo, and I was pleasantly surprised to find many related to bioinformatics. If you look in:
Bio Linux Download:-
You will see what is available as standard for biology/bioinformatics. Although Gentoo Linux is one of the most complicated distributions (every package is compiled by the system before being installed), I became a radical among radicals when seeing that much of the work already had it done, since it is the system and not I who has to compile! Have any of you ever tried to install molmol? It’s a nightmare to do it by hand! And keep BioPerl or MySQL up to date? Surely yes.
All operating systems have a greater or lesser extent of a package management system. Almost all available packages are in the binary format: the programs have already compiled, and the package system has to plant the files and little else. One not-so-known option (available in almost all Linux distributions) is the ability to use source packages: they only contain compilation instructions.
Both Gentoo Linux and FreeBSD opt for this alternative, to prepare the ‘tailor made’ packages of the system we have. The big drawback of this method is the time the system loses compiling the programs and libraries. Imagine how long it can take to install a complete system, with all the tools, programs and libraries! About a week! However, once done this I can assure you that R, BioPerl, NCBI Blast, ClustalW, etc … will work better than you think.
So, if you are radical to the core, you look for performance, and you have a real computer, I recommend Gentoo for Bioinformatics. If not, you will have to wait for the following articles, the first one on Bio-Linux (unless it persuades a fellow Debian lover to write something).
Best Laptop for Bioinformatics:-
This is also used for MacBook Air for bioinformatics
Bioinformatics on Windows:-
There is best Linux for bioinformatics. So I suggest you use always Linux for bioinformatics.